foods Review NMR in the Service of Wine Differentiation Marko Viski´c 1, Luna Maslov Bandi´c 1,* , Ana-Marija Jagati´cKorenika 2 and Ana Jeromel 2 1 Department of Chemistry, Faculty of Agriculture, University of Zagreb, Svetosimunska 25, 10 000 Zagreb, Croatia; [email protected] 2 Department of Viticulture and Enology, Faculty of Agriculture, University of Zagreb, Svetosimunska 25, 10 000 Zagreb, Croatia; [email protected] (A.-M.J.K.); [email protected] (A.J.) * Correspondence: [email protected] Abstract: NMR is a swift and highly reproducible spectrometric technique that makes it possible to obtain spectra containing a lot of information about the sample analyzed. This approach helps major components be described in complex mixtures such as wine in just one analysis. Analysis of wine metabolites is very often used to understand the impact of geographical origin or variety on wine quality. NMR is often used for tracing the geographical origin of wine. Research on NMR metabolic effects of geographical origin is of great importance as the high added value of wines results from compliance with state legislation on the protected denomination of origin (PDO) and protected geographical indication (PGI) for the administration of the appellation of wines. A review of NMR with emphasis on SNIF-NMR in the analysis of wine authenticity is given. SNIF-NMR remains a method of choice for the detection of wine chaptalization as it is the only approach which provides position-specific information on the origin of sugar in wine. However, the sample preparation step, which lacks major improvements since its conception, is strenuous and expensive, and suffers from drawbacks in terms of low sample throughput. Mainstream 1D and 2D NMR experiments provide a fast and affordable way to authenticate wine based on the geographical origin, vintage, and variety discrimination, and include a simple and non-destructive sample preparation step. With this approach, spectral data processing often represents a crucial step of the analysis. With properly performed NMR experiments good to excellent differentiation of wines from different vintages, regions, and varieties was achieved recently. Citation: Viski´c,M.; Bandi´c,L.M.; Korenika, A.-M.J.; Jeromel, A. NMR Keywords: wine; wine authenticity; geographical differentiation; NMR; SNIF-NMR; isotope analysis; in the Service of Wine Differentiation. irm-NMR Foods 2021, 10, 120. https://doi.org/ 10.3390/foods10010120 Received: 1 December 2020 1. Introduction Accepted: 5 January 2021 In the time of globalization, wine is truly one of the few products whose quality is Published: 8 January 2021 primarily determined by its uniqueness and recognizability, based on grape varieties, ori- gin, and technology. Practical wine production is no longer based on empiricism, whether Publisher’s Note: MDPI stays neu- they are mass-produced wines or small exclusive series of so-called boutique wines. A tral with regard to jurisdictional clai- prerequisite for this is the rich scientific work and the connection of research in all scientific ms in published maps and institutio- disciplines that can contribute to solving certain problems in practice. Knowledge of the nal affiliations. chemical composition of grapes and wines is essential, and analytical chemistry is the best tool to achieve this. In the 19th century, analytical methods focused on the determination of major wine components such as ethanol, organic acids, and sugars. The development Copyright: © 2021 by the authors. Li- of chromatographic techniques in the early 1900s and particularly the development of censee MDPI, Basel, Switzerland. gas chromatography in the early 1950s ushered in a new era of discovery for analytical This article is an open access article chemists [1]. The components of wine flavor were revealed by gas chromatography. As distributed under the terms and con- modern analytical methods developed, so did knowledge of wine chemistry and biochem- ditions of the Creative Commons At- istry itself. The spectroscopic methods applied in the field of analysis of wine or grape tribution (CC BY) license (https:// include a broad range of techniques, covered by atomic spectroscopic methods such as creativecommons.org/licenses/by/ atomic absorption spectroscopy (AAS) and inductively coupled plasma (ICP) and several 4.0/). Foods 2021, 10, 120. https://doi.org/10.3390/foods10010120 https://www.mdpi.com/journal/foods Foods 2021, 10, 120 2 of 21 molecular spectroscopic methods such as infrared- and ultraviolet/visible spectropho- tometry, mass spectrometry (MS), and nuclear magnetic resonance (NMR) spectroscopy. NMR spectroscopy is based on the observation that certain atomic nuclei with a non-zero intrinsic nuclear magnetic moment, when placed in an external magnetic field, can be perturbed by oscillating radio-frequency field in near resonance conditions. Perturbed nuclei undergo free induction decay and emit an electromagnetic signal with a frequency characteristic of the chemical environment at the nucleus and its neighboring nuclei. Inter- action between the magnetic moments of connected nuclei produces a coupling pattern in the signal, providing structural information about the analyte. The signal intensity, or its area, is proportional to the number of nuclei in resonance. These observations allowed for the use of NMR as one of the most advanced, sensitive, and reliable techniques used not only for the determination of molecular structure but also for quantitative analysis of real samples, purity determination, process control, medical imaging, and agricultural applications [2]. In the field of food analysis, NMR is used in compositional profiling, determination of authenticity, geographical origin of food components, and the effect of food storage and processing [3]. NMR represents the best nontargeted approach to food screening, as the majority of metabolites can be detected from a single spectrum with minimal sample destruction. 2. Application of NMR in Wine Analysis 2.1. Diversity of NMR Experiments in Wine Analysis The application of NMR in wine science has evolved significantly in recent decades. NMR can be used to detect and quantify major compounds in wine: sugars, amino acids, lactic acid, acetic acid, etc., although other more available and recognized methods with somewhat better repeatability such as NIR, HPLC, and GC are usually applied. Neverthe- less, the application of NMR has been developing in other fields of wine analysis. NMR was used to characterize the metabolome of intact grape berries [4] or berry extracts to estimate the maturity level before harvesting [5], and to assess the influence of terroir [6] and cultivation practices [7] on grape berry composition. Metabolic profiling of wines by NMR was used to group and compare wine samples [8–10]. NMR spectra of wines can show inherent genotypic variations among cultivars, which makes NMR a powerful tool for varietal classification of wines [11]. Nontargeted 1H NMR was used for recognition of Czech wine varieties, but with a very variable prediction outcome which ranged from 45% to 96% between cultivars [12]. Wine analysis by NMR can be nontargeted, as is the case of fingerprinting, which is not used for recognition of specific metabolites, involves recording multiple spectra under the same conditions, and relies heavily on multivariate data analysis to find patterns that could be used for sample discrimination. Another ap- proach is targeted analysis, which enables the identification of selected key components of wine. The targeted approach commonly employs additional structural elucidation NMR experiments such as J-res, COSY, HMBC, HSQC, and TOCSY. In one targeted approach study, Godelmann et al. (2013) grouped wines by grape varieties and production practices by identification of shikimic acid as the marker of grape variety and caftaric acid as the marker of wine oxidation. In the same study, a nontargeted approach was used to recognize patterns in the whole spectral region and distinguish between 600 samples of German wines with an excellent prediction of over 95% [13]. Targeted and nontargeted approaches were used to investigate the influence of terroir on the composition of Italian wines made from Fiano di Avellino grapes by fermentation with commercial and autochthonous yeasts. By the targeted approach, wine discrimination was based on the tracking of six different metabolites [14]. Italian Lambrusco wines were analyzed by 1H NMR fingerprinting to observe the differences between varieties [15]. Imparato et al. have used 1H NMR to identify the varieties in wine blends, but with unsatisfactory precision [16]. Concentration changes of different compounds during fermentation can be monitored, both to expand understanding of the biochemistry of fermentation and to establish new procedures for winemaking quality control [17,18]. The effect of different yeast strains on the metabolic Foods 2021, 10, 120 3 of 21 changes in monovarietal wine and their fermentative efficiency was studied by NMR with the aid of principal component analysis (PCA), partial least squares regression (PLS), and orthogonal partial least sqares regression (OPLS) at various fermentation stages [19]. Dif- ferent wine aging conditions were monitored by methods for measurement of antioxidant activity and comparison of 1D and 2D NMR spectra of phenolic species in wines [20]. NMR was used to investigate changes induced by pathological processes in downy mildew and Botrytis cinerea infected grape and in differences in the composition of produced Champagne